CausalGym: Benchmarking causal interpretability methods on linguistic tasks

{
  "doi": null,
  "year": 2024,
  "title": "CausalGym: Benchmarking causal interpretability methods on linguistic tasks",
  "venue": "Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics",
  "authors": [
    "Aryaman Arora",
    "Dan Jurafsky",
    "Christopher Potts"
  ],
  "arxiv_id": null,
  "s2_paper_id": null,
  "ingest_status": "referenced-only"
}